From 139023c19b8b6e0ebaebd20962babf4097056654 Mon Sep 17 00:00:00 2001
From: skysbird <skysbird@gmail.com>
Date: Mon, 31 Jul 2023 11:53:53 +0800
Subject: [PATCH] prepare

---
 vall_e/emb/prepare.py | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)
 create mode 100755 vall_e/emb/prepare.py

diff --git a/vall_e/emb/prepare.py b/vall_e/emb/prepare.py
new file mode 100755
index 0000000..754256a
--- /dev/null
+++ b/vall_e/emb/prepare.py
@@ -0,0 +1,34 @@
+#!/usr/bin/env python3
+
+import argparse
+import json
+import re
+from pathlib import Path
+
+
+import random
+from tqdm import tqdm
+
+
+def process_aishell(path):
+    print(path)
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--suffix", type=str, default=".wav")
+    parser.add_argument("--worker-size", type=int)
+    parser.add_argument("--dataset", type=str, default="aishell")
+    parser.add_argument("folder", type=Path)
+
+    args = parser.parse_args()
+
+    paths = list(args.folder.rglob(f"*{args.suffix}"))
+
+    for path in tqdm(paths):
+        if args.dataset == 'aishell':
+            process_aishell(path)
+
+
+if __name__ == "__main__":
+    main()